Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/sungmin-cho/deep-suite/agents-mdgit clone --depth 1 https://github.com/Sungmin-Cho/deep-suiteWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/sungmin-cho/deep-suite/agents-md)<a href="https://agentmods.dev/instructions/sungmin-cho/deep-suite/agents-md"><img src="https://agentmods.dev/badge/instructions/sungmin-cho/deep-suite/agents-md.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.01956 | $0.01956 |
| Opus 5 | $0.00978 | $0.00978 |
| Sonnet 5 | $0.00391 | $0.00391 |
| Haiku 4.5 | $0.00196 | $0.00196 |
Grade D, and why
deep-suite AGENTS.md scanned grade D with 3 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
Codex marketplace smoke test, isolated from `~/.codex/config.toml` — also from the deep-suite checkout: Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
rm -rf "$tmp_home" Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
npm test # node:test — unit + spawnSync CLI How it starts
The opening of the file, as written. The whole thing — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Deep Suite — Agent Guide
Registry and integration layer for the Deep Suite plugin family: marketplace manifests, suite-side schemas, integration guides, analysis docs, CI tooling. No plugin source code — each plugin lives in its own repo at github.com/Sungmin-Cho/deep-{name}. Treat this repo as a registry, not a monorepo.
The repositories were renamed from claude-deep-* to runtime-neutral deep-* on 2026-08-19; GitHub serves the old names as permanent redirects, so never re-create a claude-deep-* repo — doing so kills the redirect. The marketplace name stays claude-deep-suite: it is what plugin keys (deep-work@claude-deep-suite) and the ~/.claude/plugins/cache/<marketplace>/ path derive from, and renaming it would break every existing install. Codex parity comes from the mirror manifest, not a rename.
Plugins
| Plugin | Version | Description |
|---|---|---|
| deep-work | 7.3.0 | Evidence-Driven Development Protocol |
| deep-wiki | 1.10.1 | Exact worker contracts with bounded timeout fallback and journaled wiki mutation |
| deep-evolve | 3.6.3 | Autonomous Experimentation Protocol |
| deep-review | 2.10.0 | Independent Evaluator for AI coding agents |
| deep-docs | 1.7.0 | Document gardening + authoring |
| deep-dashboard | 1.5.1 | Cross-plugin harness diagnostics + suite telemetry |
| deep-memory | 1.0.6 | Cross-project semantic memory |
| deep-goal | 1.2.1 | Goal condition compiler |
| deep-loop | 1.22.1 | Loop Engineering control plane over the deep-suite |
| deep-model-router | 1.13.0 | Deterministic model/effort/review router |
Auto-generated from the marketplace manifest + each plugin's pinned
plugin.json.version. Everything outside the markers is hand-curated.
Quick Start
npm install # ajv + ajv-formats, devDeps only
npm test # node:test — unit + spawnSync CLI
npm run validate # sidecar: schema + referential integrity
npm run docs:write # regenerate marker regions
npm run docs:sync # 8 doc-sync checkers
npm run preflight # the gate: validate + docs:check + docs:sync + fixtures + test
npm run release:bump -- <plugin> <sha40> # pin → docs:write → preflight
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago Changed c44c67d44c91
- 5d ago First seen · 124 lines · 1,956 tokens per session scan D 9d4067bdecdb
deep-suite AGENTS.md is an instructions file published in the GitHub repository Sungmin-Cho/deep-suite (8 stars, last pushed 2d ago), licensed MIT. It adds 1,956 tokens to every session, about $0.0098 per session on Opus 5. A static security scan graded it D with 3 findings (reads agent configuration directories, recursive force delete, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
AmbyKit CLAUDE.md
Claude Code instructions for ambystechcom/AmbyKit, covering claude code — notes for the ambykit repo and claude-specific.
ai-auto-work CLAUDE.md
Claude Code instructions for chaohong-ai/ai-auto-work, covering gamemaker - ai native 游戏创作平台, 项目简介, ai 工作目录约定 and 按需加载（不默认装载）.
catalyst AGENTS.md
AGENTS.md instructions for coalesce-labs/catalyst, covering agents.md, what this repository is, agent philosophy, build & test and key principles.
better-codebase CLAUDE.md
Claude Code instructions for NicolasYusim/better-codebase: Read and follow AGENTS.md when developing this repository.
AIWiki CLAUDE.md
Claude Code instructions for XuebinMa/AIWiki, covering claude.md — aiwiki 项目记忆, 这是什么, 每次动手前必读（按顺序）, 写作 / 翻译用 skill（不要手搓流程） and 保持连贯的硬规则（防止跨会话的风格 / 内容漂移）.
Agentic-Ground-Rules GEMINI.md
Gemini CLI instructions for criger/Agentic-Ground-Rules, a project described as: Model-agnostic AGENTS.md templates and project instructions for AI coding agents such as Codex, Claude Code and Gemini CLI.